Transcriptome Analysis of xa5-mediated Resistance to Bacterial Leaf Streak in Rice (Oryza sativa L.)

Bacterial leaf steak (BLS) caused by Xanthomonas oryzae pv. oryzicola (Xoc) is a devastating disease in rice production. The resistance to BLS in rice is a quantitatively inherited trait, of which the molecular mechanism is still unclear. It has been proved that xa5, a recessive bacterial blast resistance gene, is the most possible candidate gene of the QTL qBlsr5a for BLS resistance. To study the molecular mechanism of xa5 function in BLS resistance, we created transgenic lines with RNAi of Xa5 (LOC_Os05g01710) and used RNA-seq to analyze the transcriptomes of a Xa5-RNAi line and the wild-type line at 9 h after inoculation with Xoc, with the mock inoculation with water as control. The results showed that Xa5-RNAi could (1) increase the resistance to BLS as expected from xa5; (2) alter (mainly up-regulate) the expression of hundreds of genes, most of which were related to disease resistance; and (3) greatly enhance the response of thousands of genes to Xoc infection, especially of the genes involved in cell death pathways, suggesting that xa5 displays BLS resistance effect probably mainly because of the enhanced response of the cell death-related genes to Xoc infection.


54
Bacterial leaf streak (BLS) is a disease caused by the gram-negative bacterial pathogen 55 Xanthomonas oryzae pv. oryzicola (Xoc) in rice. BLS is one of the most devastating 56 quarantine diseases (Niño-Liu et al. 2006) in the main rice-producing areas of the 57 world and can cause significant yield loss. BLS resistance is quantitatively inherited in 58 rice (Tang et al. 2000). In contrast to qualitative disease resistance, which is controlled 59 by single resistance (R) genes, race-specific and easily defeated by co-evolving 60 pathogens (Kou and Wang 2010), the quantitative disease resistance is driven by 61 multiple genes, generally non-race-specific and much more durable than qualitative   (Triplett et al. 2016) are also reported. Interestingly, a non-host resistance gene Rxo1 71 from maize also displays qualitative resistance to BLS in rice (Zhao et al. 2005), which 72 specifically activates multiple defense pathways related to hypersensitive response (HR) 73 against Xoc, including some signaling pathways and basal defensive pathways such as 74 the ethylene (ET) and salicylic acid (SA) pathways (Zhou et al. 2010). 75 Overexpression of a resistance protein differentially expressed protein gene 1 76 (DEPG1) that contains a nucleotide-binding site-leucine rich repeat (NBS-LRR) 77 domain results in increased susceptibility to Xoc strain RS105 and inhibition of some 78 genes related to basal defensive pathways, implying its role of negative regulation for

85
In our laboratory, a major QTL qBlsr5a conferring BLS resistance was previously 86 mapped on rice chromosome 5 (Tang et al. 2000) and further fine mapped to a 30-kb 87 interval (Xie et al. 2014). Three genes were annotated in the interval. Among them,

88
LOC_Os05g01710 was considered to be the most possible candidate gene, which

103
The present study aimed to investigate the molecular mechanism of xa5-mediated 104 resistance to BLS in rice by analyzing the effects of Xa5-RNAi on gene expression 105 regulation and transcriptional response to BLS pathogen Xoc. We revealed that Xa5- 106 RNAi could greatly alter (mainly up-regulate) the expression of many genes related to 107 disease resistance and enhance transcriptional response to Xoc, and Xa5 affects BLS   Table 1.   Total RNA of leaves was extracted using TRIzol reagent (Invitrogen, http://www.

139
invitrogen.com). First-strand cDNA synthesis was performed using PrimeScript TM RT 140 reagent kit with gDNA eraser (Takara, Japan) following the manufacturer's instruction.

141
The cDNA samples were then assayed by qRT-PCR using SYBR Premix Ex Taq 142 (Takara). The gene-specific primers used for qRT-PCR analysis are listed in Table S1.  (Table   190 2). These uniquely mapped reads were used for subsequent gene expression analysis.

191
The RNA-seq data have been submitted to the database of the NCBI Sequence Read

198
To validate the RNA-seq data, qRT-PCR (Table S1)   There were 157 DEGs (125 up-regulated + 32 down-regulated) in C2 (Table S3) 228 and 3,115 DEGs (1,202 up-regulated + 1,913 down-regulated) in C3 (Table S4), 229 respectively. The number of DEGs in C3 was ~19 times more than that in C2, MapMan analysis, suggesting that a main role of Xa5 is to regulate the expression of 251 genes related to disease resistance.

252
In C2, 78 GO terms were significantly enriched (P-value < 0.01) with DEGs (Table   253 S6), including 41 on BP, 9 on CC and 28 on MF, respectively. Almost all of the BP 254 terms were related to disease resistance, which could be also classified into several 255 groups similar to those observed in C1 except for the group of cell death.

288
As PTI has no race specificity, it is predicted to confer durable and broad spectrum Rathjen 2010). We have seen above that the DEGs in C1 were mainly enriched in four 294 groups of GO terms on BP related to disease resistance, namely, oxidation-reduction, 295 siderophore, secondary metabolite, and cell death (Table S5). Among these GO terms, 296 the former three groups belong to the mechanisms of basal defense, while the last group 297 (cell death) belongs to the mechanism of HR. Therefore, they are responsible for PTI 298 and ETI, respectively. This suggests that Xa5 regulates both PTI-and ETI-related genes.

299
It is interesting that some of the GO terms related to disease resistance in C1 were 300 also detected in the enrichment analyses in C2 and C3, but the PTI-related terms only 301 detected in C2, while the ETI-related terms only detected in C3, and there was no 302 overlap between C2 and C3 ( Figure 5). In addition, the ETI-related terms were very 303 highly significant and also the most significant in C3 (Table S7). These results suggest 304 that it is likely that the xa5-mediated BLS resistance is mainly due to the response of 305 the genes involved in the cell death pathways to Xoc infection. It appears that the 306 dominant allele Xa5 can inhibit the response of these genes to Xoc infection. Therefore, 307 no GO terms about cell death could be detected in C2.

308
Certainly, PTI-related genes may also contribute to disease resistance. But why the 309 above-mentioned PTI-related GO terms were not significant in C3? The possible reason 310 could be that the response of the genes in these PTI-related GO terms to Xoc infection 311 is similar to that to Xa5-RNAi (Figure 5), and the effects of these two types of response 312 are not additive but superimposed. Thus, since the response of the genes to Xa5-RNAi 313 has already existed in an XR line, the response to Xoc infection is masked and therefore 314 becomes undetectable.

315
In conclusion, as a component of a general transcriptional factor, Xa5 plays an 316 important role in the regulation of many genes related to disease resistance (including 317 both PTI-related and ETI-related genes) in rice. Suppression of its expression can lead 318 to defense-oriented reprogramming and thereby limit the multiplication or spread of the 319 BLS pathogen Xoc through a stronger and more direct immune response like ETI to 320 protect the plant. This defense strategy could accompany with a similar occurrence of